{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:F2MF2CZLJ7BQP7ZPNDCUUK2RPS","short_pith_number":"pith:F2MF2CZL","schema_version":"1.0","canonical_sha256":"2e985d0b2b4fc307ff2f68c54a2b517c8b0e6b4b15253949dafce0c7dbe8fd04","source":{"kind":"arxiv","id":"2601.17654","version":2},"attestation_state":"computed","paper":{"title":"Kareus: Joint Reduction of Dynamic and Static Energy in Large Model Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Jae-Won Chung, Mosharaf Chowdhury, Ruofan Wu","submitted_at":"2026-01-25T02:17:05Z","abstract_excerpt":"The computing demand of AI is growing at an unprecedented rate, but energy supply is not keeping pace. As a result, energy has become an expensive and contended resource that requires explicit management and optimization. Although recent works have made significant progress in large model training optimization, they focus on optimizing either dynamic or static energy consumption.\n  We find that fine-grained kernel scheduling and frequency scaling jointly and interdependently impact both dynamic and static energy consumption. Based on this finding, we design Kareus, a training system that pushe"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2601.17654","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-25T02:17:05Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"671d5b86c7e47b6d997917c1c97eac92767b2df3e12eb00096eb9c52e2e5f2e4","abstract_canon_sha256":"33ee9743a0516bc993089601617dde80ab0e32ee6fd944c40e5d8495ce2e64f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:09:20.765617Z","signature_b64":"ZJIrserymkTqs2DZHwKHOOhFhLOrsMFrjs/KHFqsRg9T0rXfj7LbeHH0EwYAYiw5BQPKjTGFjBiloKhGQuXNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e985d0b2b4fc307ff2f68c54a2b517c8b0e6b4b15253949dafce0c7dbe8fd04","last_reissued_at":"2026-06-12T01:09:20.764807Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:09:20.764807Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Kareus: Joint Reduction of Dynamic and Static Energy in Large Model Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Jae-Won Chung, Mosharaf Chowdhury, Ruofan Wu","submitted_at":"2026-01-25T02:17:05Z","abstract_excerpt":"The computing demand of AI is growing at an unprecedented rate, but energy supply is not keeping pace. As a result, energy has become an expensive and contended resource that requires explicit management and optimization. Although recent works have made significant progress in large model training optimization, they focus on optimizing either dynamic or static energy consumption.\n  We find that fine-grained kernel scheduling and frequency scaling jointly and interdependently impact both dynamic and static energy consumption. Based on this finding, we design Kareus, a training system that pushe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.17654","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2601.17654/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2601.17654","created_at":"2026-06-12T01:09:20.764902+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.17654v2","created_at":"2026-06-12T01:09:20.764902+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.17654","created_at":"2026-06-12T01:09:20.764902+00:00"},{"alias_kind":"pith_short_12","alias_value":"F2MF2CZLJ7BQ","created_at":"2026-06-12T01:09:20.764902+00:00"},{"alias_kind":"pith_short_16","alias_value":"F2MF2CZLJ7BQP7ZP","created_at":"2026-06-12T01:09:20.764902+00:00"},{"alias_kind":"pith_short_8","alias_value":"F2MF2CZL","created_at":"2026-06-12T01:09:20.764902+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2603.00415","citing_title":"Grid Integration of AI Data Centers: A Critical Review of Energy Storage Solutions","ref_index":123,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26103","citing_title":"AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving","ref_index":65,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05519","citing_title":"OpenG2G: A Simulation Platform for AI Datacenter-Grid Runtime Coordination","ref_index":49,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS","json":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS.json","graph_json":"https://pith.science/api/pith-number/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/graph.json","events_json":"https://pith.science/api/pith-number/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/events.json","paper":"https://pith.science/paper/F2MF2CZL"},"agent_actions":{"view_html":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS","download_json":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS.json","view_paper":"https://pith.science/paper/F2MF2CZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.17654&json=true","fetch_graph":"https://pith.science/api/pith-number/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/graph.json","fetch_events":"https://pith.science/api/pith-number/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/action/storage_attestation","attest_author":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/action/author_attestation","sign_citation":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/action/citation_signature","submit_replication":"https://pith.science/pith/F2MF2CZLJ7BQP7ZPNDCUUK2RPS/action/replication_record"}},"created_at":"2026-06-12T01:09:20.764902+00:00","updated_at":"2026-06-12T01:09:20.764902+00:00"}